Hi Mike Lowe, Thank you for your feedback. State "Deep scrubbing" occurs frequently in the system, it make user not access some data folder in ceph storage. How limited this situation?
Thanks and Regards. TienBM On Mon, Apr 22, 2013 at 10:26 PM, Mike Lowe <[email protected]>wrote: > If it says 'active+clean' then it is OK no mater what else may > additionally have as a status. Deep scrubbing is just a normal background > process that makes sure your data is consistent and shouldn't keep you from > accessing it. Repair should only be done as a last resort, it will discard > any replica's which may actually be the data you want if your primary > replica is corrupt. > > On Apr 22, 2013, at 10:07 AM, MinhTien MinhTien < > [email protected]> wrote: > > Dear all, > > - I use CentOS 6.3 up kernel 3.8.6-1.el6.elrepo.x86_64: ceph storage > (version 0.56.4), i set pool data (contains all data): > > ceph osd pool set data size 1 > > - pool metadata: > > ceph osd pool set data size 2 > > I have osd, earch osd = 14TB (format ext4) > > I have 1 permanent error exists in the system. > > 2013-04-22 20:24:20.942457 mon.0 [INF] pgmap v313221: 640 pgs: 638 > active+clean, 2 *active+clean+scrubbing+deep*; 17915 GB data, 17947 GB > used, 86469 GB / 107 TB avail > 2013-04-22 20:24:12.256632 osd.1 [INF] 1.2e scrub ok > 2013-04-22 20:24:23.348560 mon.0 [INF] pgmap v313222: 640 pgs: 638 > active+clean, 2 *active+clean+scrubbing+deep*; 17915 GB data, 17947 GB > used, 86469 GB / 107 TB avail > 2013-04-22 20:24:21.551528 osd.1 [INF] 1.3f scrub ok > 2013-04-22 20:24:52.009562 mon.0 [INF] pgmap v313223: 640 pgs: 638 > active+clean, 2 *active+clean+scrubbing+deep*; 17915 GB data, 17947 GB > used, 86469 GB / 107 TB avail > > This makes me not access some data. > > I tried to restart, use command "ceph pg repair " but error still exists > > I need some advice.. > > Thanks > > > > -- > TienBM > _______________________________________________ > ceph-users mailing list > [email protected] > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > > -- Bui Minh Tien
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
